# US 12,603,873 B2 · Dynamic one-time use knowledge-based authentication via multi-sourced private data using artificial intelligence techniques > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,603,873 B2 - Document ID: US12603873B2 - Status: Granted - Kind: B2 - Application number: 18/620,449 - Filed: 2024-03-28 - Published: 2024-10-03 - Granted: 2026-04-14 - Earliest claimed: 2023-03-29 - Assignees: Matrixed IP Holdings, LLC - Applicants: Matrixed IP Holdings, LLC - Inventors: Agasthya P. Narendranathan; James M. Dzierzanowski - Web page: https://patentgrep.com/patent/US12603873B2 ## CPC - H04L 63/08: {for authentication of entities (cryptographic mechanisms or cryptographic arrangements for entity authentication H04L9/32)} - G06F 21/31: User authentication - G06F 40/279: Recognition of textual entities - G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Exemplary systems and methods utilize unique knowledge-based authentication techniques involving private and/or recent data. Via use of the disclosed concepts, authentication is made more robust, and is hardened against compromise techniques such as those drawing from prior data breaches, public records, and the like. In this manner, simpler and more reliable authentication is achieved. ## Claims ### Claim 1 A computer-implemented method for passwordless authentication, comprising: receiving, by an authentication system, a private corpus of data associated with a user; processing, by the authentication system, the private corpus to initiate generation of a set of challenge questions; receiving, by the authentication system and over a network, a request for authentication from a user device associated with the user; reading a first record from the private corpus of data, in response to receiving the request for authentication; generating an intermediate semantic representation that associates conceptual knowledge with linguistic expressions from the first record; generating distractor content to differ semantically while remaining contextually plausible; generating distractor sentences using concepts drawn from different knowledge domains; using pre-identified tags, a genre table for the data set and semantic representation, for constructing a human-readable sentence with natural-language statements based on the first record, wherein the genre table determines vocabulary selection, sentence form and narrative framing for constructing the human-readable sentence that is contextually appropriate; maintaining domain-specific linguistic structures defining permissible vocabulary, sentence forms, and conceptual groupings used during content generation; repeating the reading and constructing steps until a plurality of human-readable sentences are constructed for each of the records in the private corpus of data; evaluating the generated content against domain rules to ensure consistency with intended subject matter prior to presentation, wherein the plurality of the human-readable sentences form a set of challenge questions; delivering, by the authentication system and to the user device via the network, a first challenge question from the set of challenge questions; responsive to receiving, by the authentication system and over the network and from the user device, the correct answer to the first challenge question, authenticating the user; and granting, by the authentication system, access to at least one of a secure computer, application or website. ### Claim 2 (depends on 1) The method of claim 1, wherein the constructing the human-readable sentence based on the first record uses at least one of pre-defined tags, multiple choice questions (MCQ) ontology or a genre configuration table that governs how questions are at least one of selected, clustered or themed. ### Claim 3 (depends on 2) The method of claim 2, wherein the private corpus of data comprises data regarding events taking place within a predetermined time frame prior to the receiving. ### Claim 4 (depends on 3) The method of claim 3, wherein the predetermined time frame is less than one week, or less than two weeks, or less than one month. ### Claim 5 (depends on 4) The method of claim 4, wherein the private corpus of data comprises at least one of credit card transactions for a credit card of the user, geolocation data for the user device, healthcare records associated with the user, or calendar appointment information for the user. ### Claim 6 (depends on 2) The method of claim 2, further comprising, responsive to the authenticating, at least one of deleting the first challenge question from the set of challenge questions or flagging the first challenge question to prevent re-use. ### Claim 7 (depends on 1) The method of claim 1, wherein the first challenge question comprises a zero-knowledge proof. ### Claim 8 (depends on 1) The method of claim 1, wherein the constructing the human-readable sentence based on the first record uses at least one of the genre table or a distractor look up table. ### Claim 9 (depends on 1) The method of claim 1, further comprising: responsive to receiving, by the authentication system and over the network and from the user device, an incorrect answer to the first challenge question, delivering to the user device, via the network, a second challenge question from the set of challenge questions. ### Claim 10 (depends on 9) The method of claim 9, wherein the second challenge question is based on more data from the private corpus of data than the first challenge question. ### Claim 11 (depends on 10) The method of claim 10, further comprising flagging, by the authentication system and in the set of challenge questions, the first challenge question and the second challenge question to prevent reuse of the first challenge question and the second challenge question. ### Claim 12 (depends on 11) The method of claim 11, further comprising deleting, by the authentication system, the flagged challenge questions from the set of challenge questions. ### Claim 13 (depends on 1) The method of claim 1, wherein the first challenge question is a multiple-choice question. ### Claim 14 (depends on 1) The method of claim 1, further comprising: receiving, by the authentication system and over the network, update data associated with the user; adding, by the authentication system, the update data to the private corpus of data; processing, by the authentication system, the update data to generate additional challenge questions; and adding, by the authentication system, the additional challenge questions to the set of challenge questions. ### Claim 15 (depends on 1) The method of claim 1, further comprising annotating, by the authentication system and in the set of challenge questions, each challenge question with a use-by date by which the challenge question should be used or discarded. ### Claim 16 (depends on 15) The method of claim 15, wherein each record in the private corpus of data has a generation date, and wherein the use-by date is a predetermined number of days from the generation date of the record in the private corpus from which the associated challenge question was generated. ### Claim 17 (depends on 1) The method of claim 1, further comprising deleting, by the authentication system and from the set of challenge questions, at least one challenge question having a use-by date prior to a current date.